PROBLEM OF PREDICTING RISK OF CORROSION OF STEEL IN CHLORIDE CONTAMINATED CONCRETE. PROCEEDINGS OF THE INSTITUTION OF CIVIL ENGINEERS

Corrosion of reinforced steel in concrete structures is a widespread international problem, with affected structures frequently showing signs of severe deterioration after only a few years of service. Repair of these structures is very expensive and the costs incurred have imposed a huge financial burden on the economies of several countries. Chlorides are generally believed to be one of the primary causes of corrosion of steel embedded in concrete. A direct method of predicting the corrosion risk involves measuring the chloride content of concrete and comparing it with a threshold value, which is the chloride level required for corrosion initiation. Three major approaches have been followed in order to establish this threshold chloride content by measuring: total (acid-soluble) chloride content which is proposed in ENV 206 and is the current UK technique; water-soluble chloride content which is widely used in North America; the chloride/hydroxyl ion concentration ratio in the pore solution used primarily by researchers. However, it is generally recognized that it is only the chloride ions in the pore solution (i.e. free chloride), rather than the total chloride, that contribute to the corrosion of embedded steel. In view of this, the Paper examines the rationale behind the above three approaches. It is shown that, although the total chloride content can be determined readily, it may be a poor indicator of corrosion risk since no allowance is made for mix composition or source of chloride contamination, both factors of which are found significantly to affect the level of chloride ions in the pore solution. The work described in this Paper also suggests that the water-soluble chloride content may be an inappropriate parameter to relate to corrosion risk since it provides only an approximation of the free chloride content over a limited range of mixes. The free chloride/hydroxyl ion ratio is a more realistic indicator of corrosion risk, but there are practical difficulties associated with its determination in actual structures. The Paper describes an alternative procedure to the above which allows the free chloride content of concrete to be determined by way of the measurement of total chloride, using empirical relationships derived from results obtained by laboratory extraction of free chloride. The aim of introducing this new method is to rationalize the collection of relevant chloride content data, thereby improving the prediction of the risk of corrosion in reinforced concrete structures. To establish suitable critical limit for free chloride content, it is necessary to test concrete and steel from actual structures, and a plea is made for the facilities to accomplish this. (Author/TRRL)